J. Chem. Eng. Data 1001, 36,277-280
Nr = salt mole fraction In the ethanol-water-potassium acetate -liquid phase = (mol of potassium acetate)i(mol of potassium acetate mol of ethanol mol of water) N,' = salt mole fractbn in the ethanol-water-salt mixtures liquid phase = (mol of potasslum acetate mol of sodium acetate)/(mol of potassium acetate mol of sodium acetate mol of ethanol mol of water) P = system pressure (mmHg) P o , = vapor pressure of pure component i (mmHg) P = corrected vapor pressure of component i(mmHg) T = temperature liquid phase ( O C ) x, = mole fraction of component i In the llquid phase, calculated on a salt free basis = (mol of ethanol or water)/(molof ethanol mol of water) y, = mole fraction of component i in the vapor phase = (mol of ethanol or water)/(mol of ethanol mol of water) a, = relathre volatility of ethanol in the presence of the salt y, = activity coefflclent in the liquid phase of component i
+
Literature Cited
+
+
+
+
277
Gorhan, A. U S . Pat. 1879847,1932 Chem. Absi?. 1988,27, 369. Furter, W. F. ACS 180th Natbnai Meeting, CMCago, IL, Sept 1970. Meranda, D.; Furter, W. F. Can. J . Chem. Eng. 1966, 14 (5), 298. Caok, R. A.; Furter, W. F. Can. J . Chem. Eng. 1968, 48, 119. Costa Novella. E.; Moragues Tarrasb, J. An. R . Soc. Esp. Fls. Wm. 1952. 486,441. Meranda, D.; Furter, W. F. AIChE J . 1971, 17, 38. Michalowski, S.; Mondeja, D. Cent. Azucar 1977, 4 (3),25. Johnson, A. I.; Furter, W. F. Can. J . Chem. Eng. 1965, 43, 356. Sabarathlnam, P. L.; Andlappan, A. N.; Lakshmanan, S. M. Chem. I d . D e v . 1975, 9,27. Schmitt, D. Ph.D. Thesis, Universlty of Karlsruhe, Germany, 1979. Walas, S. M. h s e Equilibnle in Chemical Engimwhg; B u t t e " : London, 1985. Gmehling, J.; Onken, U.; Ark, W. Vapor-UqUM Equlllikhrm &ita C&xtion; MCHEMA: Frankfurt, 1981; Vol. I l l . Jaques. D.; Furter. W. F. AIChE J . 1972, 18, 343. Natarajan, T. S.;Srinivasan, D. J . Chem. Eng. Data 1900, 25, 215.
+
*,
+
+
Densities and Viscosities of NH,Br-NH,
and NH,I-NH,
Systems
Hideki Yamamoto' and JunJlTokunaga' Department of Chemical Engineering, Faculty of Engineering, Kansai Universi@, Yamate-cho, Suita, Osaka, 564 Japan
Dendtles of the NH,Br-NH, and NH,I-NH, systems in ilquM phase have been measured over a wlde range of temperatures and concentrations. Densities of saturated soiutions of the NH,Br-NH, and NH,I-NH, systems have also been measured over the temperature range from 0 to 76.5 O C and from 0 to 69.5 O C , reswlvely. The accuracy of this measurement was *0.1% . Observed densities were expressed as a functlon of temperature and concentratlon by means of polynomial equatlons. These equatlons could calculate values to an accuracy within fO.5% of the observed data. For the sake of the viscometry of these systems, a failing-body viscometer with use of a He-Ne laser was assembled. The vlscoslties of NH,Br-NH, and NH,I-NH, systems were measured In the ranges of mass percent of ammonium halide from 10.0% to 59.8% and from Q.81% to 70.1% under the wlde temperature range respectlveiy. The accuracy of these vlscometry was withln *1.2% devlatlon aaainst standard liquid to calibrate thls vlscometer. Introduction Very little information for the physical properties of ammoniated salts in the liquid phase is available for the concentrated solution. Most of the Investigations of ammonium salts In the literature involve mainly aqueous solutions. Although dilute aqueous solutions have greatly contributed to the development of physical chemistry, the behavior of concentrated nonaqueous electrolytic solutions has not yet been completely investigated. Progress in theoretical descriptions of these ammoniated salts requires the availability of experimentally determined data. I n recent years, some ammoniated salts were used as working system for chemical heat pumps and energy storage ( 1 , 2). However, most of the studies in the literature involving ammoniated salts have been carried out at or near normal boiling point to avoid experlmentatlon at hlgh pressures. Some physical and thermodynamic properties of NH46r-NH, and 002 1-956819 1I1 736-0277$02.50/0
NH41-NH, (Le.vapor pressures, solubilities, specific heats, and heats of mixing) were already measured in this laboratory (3, 4 1. I n this paper, the densities in the wide temperature and concentration ranges were measured by using the special equipment that was made to facilitate measurements at high pressure. Observed densities were expressed as the function of temperature and concentration, using the method of ieastsquares. Furthermore, the viscosities of these systems were measured by using the failing-body viscometer, which was specially assembled for the measurement of the solution under high pressure. This paper is concerned with the densities and viscosities of the NH,Br-NH, and NH,I-NH, systems. Matedab. Ammonium iodide (NH,I) and ammonium bromkh (NH,Br) from Wako Pure Chemical Industries, Ltd., were of guaranteed reagent grade and were specified as the pure grade, having minimum purities of 99.5%, and used without further purification. The powdered crystal was thoroughly dried at 100 OC and stored over silica gel in a desiccator. Ammonia gas of 99.99% purity was provided by Seitetu Kagaku Co., Ltd. The standard liquid for calibration viscometers was used in order to examine the accuracy of the falling-body viscometer from Showa Shell Sekiyu Co., Ltd. The range of viscosities for the standard liquid was 0.1-5.0 Pa s.
Apparatus and Procedures Densny Measurement. A schematic diagram of the experimental apparatus for density measurements is shown in Figure 1. The two vessels (Iand J) are made of pressureresistant glass (up to 2 MPa): one is a 20-mL vessel for measuring the volume of liquid ammonia (ammonia vessel), and the other is a 100-mL vessel for determining the densities (reaction vessel) that can be agitated by a magnetic stirrer (6). Two vessels were immersed in a constant-temperature water bath separately, and its accuracy was within f0.05 OC. The pressure in the reaction vessel was measured by a strain gauge transducer (A) with a accuracy of 0.1 % of full scale (2 MPa). 0 1991 American Chemical Society
270 Journal of Chemhl and Engineering Data, Vd. 36, No. 3, 1991
I. Schematic &gam of the apparatus: (A) disltel pressure gauge: (E) maptic stirrer: (C)recorder; (D) Bovdon gauge: (E)water bath; (F) thermocouple; (0)selective switch: (HI a"onla vessel; (K) vacuum pump; (L) microscope. FIQWO
U
The volume of the solution and the liquid ammonia were measured by microscopes (L) within f0.02% of the full volume. The temperatures in the vesseles and pipe arrangement were measured by chromel-alumel thermocouples (F) corrected by a standard mercury thermometer. The temperature in the vessel was kept within f0.05 OC. After a weighed amount of solid material (maximum 0.04 kg) was charged in the reaction vessel, the pipe arrangement and two vessels were kept at 80 OC under vacuum for 2 h in order to remove the moisture from the system. Then, these vessels were cooled to normal temperature. Chilling the ammonia vessel allowed liquid ammonia to be introduced into the ammonia vessel from an ammonia cylinder. Thereafter, the ammonia vessel was kept at 25 OC. Then, the Wqukl ammonia was transferred to the reaction vessel from the ammonia vessel of different temperature so as to dissolve ammonium haw in the reaction vessel. The density and concentration were determined from the weight of the solvent and solute and also from the volume of solution in the reaction vessel. The quantity of liquid ammonia transferred to the reaction vessel from the ammonia vessel and the volume of solution in the reaction vessel were measured by reading the scales of these two vessels with use of a microscope, respectively. I n the case of a saturated solution, most of the ammonium halide dissolved in the liquid ammonia. Then, all valves were closed, and the temperature of reaction vessel was raised slowly at the rate of 0.5 OC/min to dissolve the solute. When the solution of ammonium halide was around saturation, the rate was decreased to 0.2 OC/min or less. The saturated temperature were read at the point where the last crystal disappears. The liquid level in the reaction vessel was also noted immediately. Then, the solution in the reaction vessel was cooled gradually from a temperature higher than the saturated point, and the temperature at which the first crystal appears was recorded. The average of the these two densities was determined as the density of the saturated solution. I n order to prevent the condensation of ammonia gas, a heating coil was rolled around the pipe arrangement and was kept at constant temperature. The correction for the weight of ammonia in the pipe arrangement was calculated by using the equation of state for ammonia gas by Meyer and Jessup (5). V/scos/ty Meauur8mont. As the NH,Br-NH, and NH,I-NH, systems are high-pressure and corrosive solutions, viscosity measurements of these solutions are not so easy. For this purpose, the faillng-body viscometer with use of a He-Ne laser was specially assembled in our laboratory. Figure 2 shows the experimental apparatus of the viscometer. The diameter and height of the cylinder were 1.270 X lo-* m and 0.45 m, respectively. The two kghts of the He-Ne laser (A) were inserted into the cylinder from the two pressure-resistant glass windows (C). The length of the two windows was 2.0 X lo-* m. Then, the light coming from the cylinder was detected by the photo-
U
Fbure 2. Schematic diagram of the experimental apparatus: (A) M - ~ laser; e (E) water bath: (C)optical wlndow; (D) optical fiber; (E) detector circa; (F) cUgttal oscilkscope;(GIelectric "ter; (HI cylinder: (I) water bath; (J) falling body; (K) collector; (L) mount; (MI optical bench; (N) power supply; (0)circulator; (P) Ranthanet magnet.
F W e 3. Glass vessei for the volume measurement of a falling body.
transister through the optical fiber (D),the diameter of which was 1 .O X 1 O3 m. The signals detected by the phototransister were analyzed using a digltai storage oscUloscope and a dlgltal electric counter. I f the falling body passed through the glass window, laser light was intercepted by the falling body in the cylinder, and the time required for passing between the two glass windows midbe measured. The terminal velocity of the falling body was calculated from the length between the two glass windows and the passing time. The digital resolution of the time measurement was 0.1 ps. The upper part of the sealed cylinder could be charged with four falling bodies (each body with different density) at the same time. After the experiment, the falling body at the bottom of cylinder was pulled up by using a Ranthanet magnet collector seated with Teflon. FurthemKKe,the cylinder was covered wlth a constant-temperature water bath, with an accuracy within f0.05 OC. Figure 3 shows the glass vessel for the measurement of the volume of the falling body. At first, this glass vessel was filled with water at 25 OC up to a fixed line. Thereafter, the falling body was placed in the apparatus, and the glass vessel was again filled with water up to the fixed line. The volume of the falling body was calculated from the difference of its weight and the mass difference of this vessel. For the experiment, after a weighed amount of solid material was charged In the stainless steel cylinder, it was kept at 80 OC under volume for 2 h In order to remove the moisture from the system. Then, these vessels were cooled to normal temperature. ch~lingthe ammonia vessei (its v o h w was lo3 m3) allowed liquid ammonia to be introduced Into the ammonia vessel from the ammonia cylinder. Thereafter, the ammonia vessel and cylkbr were kept at 40 and 0 OC, respectbe, and ilquid ammonia was transferred to the cylinder from the ammonia vessel. The quantity transferred to the cylinder from the ammonia vessel was determined from weight difference of the ammonia vessel. Its accuracy were within 0.001 g. The mass
Jownal of Chemical and Engineering Dsta, Vol. 36,No. 3, 199 1 279
Table I. Densities (p) of the NH4Br-NH, System at Various Temperatures D , kg/m3 0 "C 10 "C 20 "C mass % of NH4Br 681 706 695 9.8 769 755 19.7 780 854 843 30.1 865 941 927 40.5 948 1043 1033 50.2 1051 1159 1147 60.8 1168
30 "C 669 743 829 916 1023 1137
Table 11. Densities (p) of the NH41-NH, System at Various Temperatures D. kg/m3 mass % of NH41-NH3 0 "C 10 "C 20 "C 700 694 10.1 710 785 776 19.8 794 881 871 30.4 890 986 977 40.2 993 1101 1094 49.9 1109 1230 1223 60.3 1239 1349 1326 69.8 1373
30 O C 681 764 862 968 1084 1211 1302
Table 111. Coefficients a;. ~.. b;,for Least-Sauares ReDresentations of Density NH4Br-NH3system bo bl
b2 b3
a0
a1
02
a0
-1.38182 X 7.77802 X 10" -6.39803 X 10" 1.78457 X -1.52249 X lo4
-2.6147 X 10" -1.20299 X 10" 1.13774 X lo-' -3.17576 X 10" 2.68685 X lo-"
0.638905 6.50616 X 10" 5.72231 X 2.97606 X 10"
percent of the solution was determined from each weight. The faliingbody viscometer is a popular and simple vlscometry apparatus. By utilizing Stokes equation for a falling body under the influence of gravity in an infinite Newtonian fluid medium, the time difference (At) for passing through the two points in the cylinder, the densities of the falling body and fluid (ps,pf), and the geometry constant (K)were needed to calculate the Viscosity from the following relation: c1
= K(P, - Pf)At
(1)
The geometry constant K was expressed as a function of the form of the falling body and diameter of cylinder. K was determined from some standard liquids for calibrating viscometers. The reproducibility of geometry constant measurements was within f1.5%. The viscosities were calculated from eq 1. For confirmation of whether or not the observed velocity was the terminal velocity, two falling bodies were tested in the same fluM. I f the density (pf)of the solution, obtained from eq 2, was Kl(pSl
- &IAf
1
= K2(pS2
- Pf)Af2
(2)
in agreement with the density determined from its initial weight and-mass within a accuracy of f0.5% , the observed viscosity was adopted for use in this experiment.
50 "C 637 714 807 891 997 1114
50 "C 649 737 837 943 1061 1192 1257
40 O C 666 750 851 957 1073 1203 1280
NHJ-NH, system
0.638412 6.92786 X 7.18433 X 10" 3.64899 X
b4
40 "C 653 726 816 903 1014 1125
a1 -1.35588 X 1.46554 X lo4 -9.30585 X 10" 2.19404 X -1.70813 X 10"
a2
-3.45319 X 10" -1.25946 X lo4 7.50838 X lo4 1.55015 X lo4 1.07121 X lo-"
Table IV. Densities of Saturated Solutions for the NH,Br-NH. and NH,I-NH* Systems NH4Br-NH, system NH41-NH3system temp, concn,' density, temp, concn,' density, "C % kg/m3 "C % kg/m3 0 77.05 1526 0 68.58 1268 1531 77.39 1269 6.1 1.7 68.70 1534 10.0 77.63 6.0 69.01 1271 1540 15.5 77.97 11.6 69.42 1274 1545 19.9 78.26 16.3 69.89 1277 1551 1279 24.8 78.60 20.1 70.25 1558 30.2 79.01 26.0 70.87 1282 71.31 1564 79.38 1284 34.9 29.9 1570 38.7 79.71 33.0 71.68 1286 1573 1287 40.9 79.90 34.5 71.86 1576 35.2 71.92 1288 43.5 80.14 1579 1289 45.2 80.30 40.9 72.08 1587 1290 50.4 80.82 44.6 72.18 1595 51.7 72.39 1292 55.5 81.37 72.60 1601 1295 59.0 82.38 58.5 1609 63.7 72.76 1296 63.9 82.38 1616 82.95 1298 68.2 69.6 72.95 76.5 73.18 1618 1302 69.4 83.11
Mass percent of sr-ite (NH4Bror NH41).
Rerun and discussion D"y. The densities of the NH,Br-NH, and NH,I-NH, systems are shown in Tables I and 11. All density data were fitted to a quadratic correction, given as eq 3 and 4. 2
p
=
1 0 0 0 ( ~a, T') 1-0
(3) 100
(4)
Values for a,and b,are listed In Table 111. Smoothed values calculated from this equation agreed with the observed data within f0.5%. Table I V shows the densities of the saturated solution for the NH,Br-NH, and NH,I-NH3 systems. The ac-
10'
Re
102
Id
1-1
Flguro 4. Relatlon between geometry constant K and Re.
curacy of the density measurement was estimated within f0.1% , judging from the accuracy of microscope. Vlscorlty. Flgure 4 shows the relation between the geometry constant (K)and the Reynolds number. These constants
200 Journal of Chemlcal and Engineering Data, Vol. 36,No. 3, 199 1
Table V. Viscosities ( p ) of the NH,Br-NHs System at Various Temperatures
Table VI. Viscosities ( p ) of the NH,I-NHs System at Various Temperatures
of NH,Br
10.0 "C
20.0 "C
f i . Pa s 30.0 "C
40.0 "C
50.0 "C
of NHJ
10.0 "C
10.0 21.7 30.2 40.9 49.5 59.8
0.208 0.316 0.485 0.781 1.390 3.535
0.191 0.280 0.398 0.682 1.163 2.843
0.177 0.252 0.351 0.595 1.011 2.310
0.156 0.224 0.308 0.517 0.877 1.810
0.140 0.210 0.280 0.452 0.755 1.461
9.81 19.3 30.5 40.2 49.9 60.0 70.1
0.195 0.244 0.345 0.490 0.811 1.360 4.625
mass %
t
mass 90
NH4Br-NH3 %stem
L \
20.0 "C 0.178 0.222 0.312 0.438 0.710 1.382 3.720
P , Pa 30.0 "C 0.162 0.201 0.282 0.395 0.634 1.220 3.243
40.0 "C 0.146 0.182 0.250 0.348 0.557 1.118 2.800
50.0 "C 0.138 0.170 0.230 0.315 0.498 0.972 2.430
temperature for solutions give straight lines being nearly parallel. The accuracy of this fallingbody viscometer was wlthin f 1.2% , judging from the measurements of standard reference materials.
Concluslon The density and viscosity of NH,Br-NH, and NH,I-NH, were measured over a wide range of temperatures and concentrations. The experimental apparatuses used in this paper were special equipment assembled to facilitate measurements at high pressure.
Nomenclature
Temperature I'CI
Figure 5. Viscosities of the NH,Br-NH,
system.
- 5
a = coefficient of least-squares equation b = coefficient of least-squares equation K = geometry constant defined in eq 1 (cm2/s2) t = time difference (s) T = temperature ("C) X = mass percent of solute (%) Greek Letters p = viscosity (Pa s) p = density (kg/m3) Subscripts f = fluid
s = solid 1 = number of falling body 2 = number of falling body ReGISlry No. NH3, 7664-41-7; NH,Br, 12124-97-9; NH,I, 12027-06-4. "0
10
20
30
40
50
Temperature I'C 1
Figure 6. Viscosities of the NH,I-NH,
system.
( K )were determlned from eq 1, as shown in Figure 4. These data were constants under Re = 3, and the experiment was
carried out in the range of Re < 2. The observed viscosities are listed in Tables V and VI, and Figures 5 and 6 show the temperature dependence on the viscosity of NH,Br-NH, and NH,I-NH3 systems. The plots of logarithms of the viscosity against the reciprocal of the absolute
Llterature Cited (1) MaclMen, M. 0.; Klein, S. A. Soler Enefgy 1989, 37 (5). 473-482. (2) Yamamto, H.; Sanga, S.; Tokunaga, J. Ind. Eng. Chem. Res. 1987, 26, 2389-2393. (3) Yamamoto, H.; Tokunaga, J.; Sanga, S. J. Chem. Eng. Date 1986. 3 7 283-285. (4) Yamamoto, H.; Sanga, S.; Tokunaga, J. J. Chem. Eng. Data 1988. 33, 381-385. (5) Meyer, C.H.; Jessup, R. S. Refrig. Eng. 1924, 7 7 , 348. I
Received for review May 23, 1990. Revised October 2, 1990. Accepted November 26, 1990.
